Metabolites of arachidonic acid (AA) are potent local mediators that control a wide range of functions under both normal and abnormal conditions. Cyclooxygenase-2 (COX-2), a key enzyme in the metabolism of AA within the central nervous system (CNS), is the therapeutic target of many highly self-prescribed over the counter anti-inflammatory/analgesic drugs. It is expressed under normal conditions in neurons of the hippocampal CA3 region and layers 2/3 of the cerebral cortex. Relevant to this proposal, convulsive seizure activity induces upregulation of COX-2 expression in these regions as well as in neurons of the dentate gyrus (DG). This raised the possibility that COX-2 may serve a neuromodulatory function in epilepsy, a brain disorder characterized by spontaneous seizures. Indeed, evidence from numerous studies, including preliminary and published results from the PI's laboratory, supports the notion that products of neuronal COX-2 activity may possess antiepileptic properties. This central hypothesis forms the basis for the proposed studies of this proposal. Studies in Aims 1 and 2 will test the effect of targeted transgenic (TG) overexpression of neuronal COX-2 on acute seizure threshold and epileptogenesis, the process by which the brain becomes epileptic. In both Aims, it is hypothesized that region-specific upregulation of COX-2 expression will reduce incidence of acute convulsive seizures and epileptogenesis. To test this, studies in Aim 1 will examine the consequences of targeted COX-2 overexpression in neurons of the DG or CA3 regions of the hippocampal formation. Complementary studies in Aim 2 will further test the effects of targeted COX-2 overexpression in both the CA1 hippocampal neurons and excitatory layer 2/3 neurons of the cortex. A novel TG mouse line that permits Cre-dependent upregulation of COX-2 will be used in each of these Aims. These mice will be crossed with TG mice that use neuron-specific promoters to target Cre expression to neurons of each of the 3 desired brain areas. An added advantage of the TG Cre line to be used in Aim 2 is that it will permit tamoxifen-induced expression of Cre. Thus both special and temporal control of COX-expression will be possible. Double TG mice from each of these crosses will be tested in models of acute seizures and epileptogenesis. Aim 3 studies will explore the mechanisms that control the level of neuronal COX-2 expression from its native gene. Specifically, it is posited that loss of TIA-1, an mRNA binding protein that is known to bind to and suppress COX-2 mRNA translation, will lead to enhancement of COX-2 expression in neurons. A targeted mutant mouse line will be employed to test this hypothesis in pure cortical cultures of neurons and in animal models of epilepsy. Studies in the three Aims of this proposal may suggest novel ways (e.g., Gene therapy) by which epileptogenesis or established epilepsy may be arrested or perhaps even reversed.